One photon. One bent molecule. Everything else is chemistry.

Inside every photoreceptor, folded into a protein called opsin, sits a small molecule named 11-cis retinal. It is a derivative of vitamin A, and it has a kink in it. When a photon of the right energy is absorbed, that kink straightens. The molecule becomes all-trans retinal.

That is it. That is the entire light-dependent event in human vision. Nothing downstream of it needs a photon, nothing upstream of it does anything but deliver photons to the right place. George Wald took a Nobel prize for working out that the chemistry of sight comes down to a vitamin A derivative changing shape, and the finding has not needed revising.

The consequence runs further than it looks. Because retinal is a vitamin A derivative, dietary vitamin A deficiency is a direct failure of the light-absorbing apparatus itself. Rods carry far more pigment than cones, so they fail first, and impaired dark adaptation is the earliest sign. Night blindness from vitamin A deficiency remains one of the leading preventable causes of blindness in the world, and it is this molecule that explains why.

Here is the part that surprises anyone who assumes the eye works like a camera sensor. In darkness, cyclic GMP holds cation channels open. The photoreceptor sits depolarised, current pours across its membrane, and it releases neurotransmitter continuously. That standing current has a name: the dark current.

Light destroys the cyclic GMP, the channels close, the cell hyperpolarises, and transmitter release falls. Photoreceptors do not fire when they see light. They stop.

Follow the metabolic consequence, because it runs through everything else on this site. Maintaining the dark current means running ion pumps hard, and running ion pumps hard means consuming ATP. The expensive state for a photoreceptor is darkness. These cells are among the most energy-demanding in the body, their inner segments are packed with mitochondria to pay for it, and their peak demand is at night.

Now recall from stage 04 that the cones at the centre of the fovea sit in a zone with no capillaries at all, fed only by diffusion from the choroid behind them. The most energy-hungry cells in the eye are supplied by the least direct route available to them. That single mismatch is the structural reason so much of what goes wrong with central vision goes wrong precisely there.

All-trans retinal cannot absorb another photon. Before the photoreceptor can respond again, the molecule has to be straightened back into its 11-cis form, and the photoreceptor cannot do that itself. It is transported out to the retinal pigment epithelium, re-isomerised there through a chain of enzymes including RPE65, and returned. That round trip is the visual cycle.

Its speed is something you have felt. Walk from daylight into a dark cinema and the first minutes of recovery are cones regaining sensitivity. Several minutes in, the rods overtake them, a transition visible on any dark adaptation curve as the rod-cone break. Full rod sensitivity takes closer to half an hour. Your pupils finished dilating in under a minute. What you were waiting for was chemistry.

When the enzymes in that loop fail, vision fails with them. Inherited mutations in RPE65 cause a severe childhood retinal dystrophy, and that specific defect became the target of the first gene therapy approved for an inherited retinal disease. The pathway described in these four paragraphs is not background. It is where the treatment had to aim.

The light-sensing part of a photoreceptor is a stack of membrane discs, and it is not permanent. New discs are assembled at the base of the outer segment and old ones are shed from the tip, so the whole apparatus is replaced roughly every ten days. The shed material does not simply dissolve. The retinal pigment epithelium engulfs and digests it, on a daily circadian schedule, for every photoreceptor it serves.

Consider that workload across a lifetime. Each RPE cell supports dozens of photoreceptors and processes their discarded outer segments every day for eighty years, and it is a cell that does not divide. The residue accumulates as lipofuscin. Beneath it, Bruch's membrane, across which every nutrient and every waste product must diffuse, thickens and stiffens.

That is the ground on which age-related macular degeneration develops, and it is why the RPE, rather than the photoreceptors themselves, is where the story usually starts.